In a related technology, Cloud computing is known in which machine resources can be obtained on an as needed basis. The Cloud computing is expected to be more and more used among various users because cost can be reduced due to sharing of resources among multiple users.
Further, when the Cloud computing will be mainly used, the Cloud computing will be widely used in a manner such that operations are performed on a tenant (e.g., a system that is provided for each of Cloud users) basis due to automatic operation in Cloud. To perform the automatic operation in the Cloud, various handling methods are generated in advance, and the automatic operation is performed using one of the generated handling methods. Here, the “handling method” refers to, for example, various rules, which are used to perform the automatic operation, describing how to handle in respect to various failures and problematic items (events).
As an example of the automatic operation using the handling method, there are a method of preparing a handling method which including a set of handling methods which may be desired for the operations, and an operating method using a handling graph. The “operating method using a handling graph” refers to a method in which cause-and-effect relationships related to an event occurred in a system are visualized using the handling graph, and handling is performed on the event which is observed in the actual system by using the handling graph.
The handling graph is generated by, for example, linking the cause-and-effect relationships between the handling history records in the past and the corresponding events by using a predetermined method which includes a method in which a user determines whether the cause-and-effect relationship exists or a method in which determining whether the cause-and-effect relationship exists is based on statistical data. Further, recently, in order to promptly generate a handling method to perform the automatic operation for a tenant which has changed a configuration or has added a new configuration, which may influence a system operating method, there is a known method in which a pair of the cause-and-effect relationships of the handling graph is divided and managed. In a specific handling method, for example, a handling element figure is generated or a handling graph for a new application is generated. The “handling element figure” refers to a figure in which the handling draft graphs of all the applications on the Cloud are divided for each of the sets of the cause-and-effect relationships, and the groups based on the application characteristics having the cause-and-effect relationships as elements (handling graph elements) are managed. Further, the handling graph for a new application acquires a group of the handling graph elements corresponding to the characteristics of the new application, and connects the cause-and-effect relationship.
Further, a method is known in which in an information system, the handling cases of the failures occurred in the past are registered as failure handling knowledge, so that when a failure occurs, a handling method is recommended based on a symptom of the failure using the acquired failure handling knowledge. Further, a method is known in which job execution cases in the past are classified and stored as know-how in the order corresponding to the job procedure, so as to assist based on the job execution cases in the past in which failure control measures when a failure occurs are stored in order to effectively execute a new job (see, for example, Japanese Laid-open Patent Publication Nos. 2010-72834 and H8-314751).